Technical Reviews
Current Research Projects
- SOC and SOH Dependence in Lithium-ion Battery State Estimation
Objective: To present a generalized method for incorporating state
of health and state of charge estimators into a combined
framework
Summary:Combining state of charge (SOC) and state of health (SOH)
techniques together to create a self-adaptive battery monitoring system. The proposed method
is able to utilize SOH predictions to update the SOC estimator in order to minimize modeling
errors due to capacity loss and cell degradation.
- State of Charge Estimation for Electric Vehicle Batteries under an Adaptive Filtering Framework
Objective:To develop a method for battery state of charge estimation
under dynamical discharge.
Summary:Presented an Unscented Kalman filtering-based approach to perform SOC estimation under dynamic loading
conditions. The proposed approach is able to self-adjust the battery model parameters and provide the SOC estimation
with a known level of confidence.
- Implementation of State of Charge Estimation for Lithium-ion Battery Balancing
Objective:Investigate the effect of cell imbalance on battery capacity
fade; Propose an implementation of state of charge estimation
into battery balancing.
Summary:Presented a state-ofcharge
estimation-based method to investigate the effect of cell imbalance on battery capacity fade. The proposed
method is able to detect cell imbalance more accurately and keep all the cells operating within safety limits.
- Early Detection of Fan Bearing Faults using Acoustic Emission Signals
Objective:Develop a method that selects optimal model parameters
for the sequential probability ratio test to detect faults
Summary:Presented the use of the sequential
probability ratio test for the detection of statistical changes against a null hypothesis/healthy data at the earliest possible
time. The proposed method is able to extract features in a statistical way from sensor signals that are indicative of and
sensitive to bearing degradation.
- Fault Simulation and Diagnostics using System-Level Signals by Dynamic Modeling
Objective:To evaluate the use of system level signals for fault detection
in mechanical equipment.
Summary:Presented the use of
system-level signals to perform component-level diagnosis. The proposed method is able to detect and identify
component-level faults in a complex system, such as a wind turbine, with a greatly reduced number of monitoring
sensors.
- Diagnosis of Rolling Element Bearing Fault in Bearing-Gearbox Union System using Wavelet Packet
Correlation Analysis
Objective:Extract bearing fault features for bearing-gearbox union
systems
Summary:Presented a novel feature extraction method based on wavelet packet transform,
correlation analysis, and envelope analysis. The proposed method is able to extract fault features of the rolling element
bearing from the masking faulty gearbox signals.
- PHM Framework for the Real-Time Assessment of Analog Circuits
Objective:Introduce a PHM framework for analog circuits with a focus
on:
• early fault detection and isolation
• failure prognostics
Summary:Presented a systematic approach to
perform online soft-fault diagnosis and prognosis on actual analog circuits subjected to component variations within
a standard tolerance range. The proposed approach is capable of detecting and isolating faults, and predicting circuit
failure in real-time, and thus, can be used to evaluate the reliability of electronic systems during field operation.
- PHM Software Suite
Objective:To develop a PHM software suite that can be implemented
on-board and in real-time
Summary:Presented a generic, modular, flexible, and interoperable software architecture that
effectively and efficiently integrates various PHM routines. The proposed software can be implemented on-board
and in real-time with easy deployment and maintenance/update.
- Degradation Mechanisms of Inductively Loaded Field-Stop Trench-Gate Insulated Gate Bipolar Transistorsunder Power Cycling
Objective:• Determine failure mechanisms of power cycled IGBTs by
physical analysis
• Collect data for future data-driven techniques development
for anomaly detection and remaining useful life estimation
Summary:Presented a study of the degradation of FS-TIGBTs under inductive load with
different switching frequencies, duty cycles, and temperature swings. The study is able to identify the effects of
different loading conditions on different failure mechanisms of FS-TIGBTs through multiple failure analyses,
including cross-sectioning, scanning acoustic microscopy (SAM), scanning electron microscopy (SEM), and others.
- Anomaly Detection for IGBT under Power Cycling using Statistical Approach
Objective:To identify, update and implement an algorithm to detect the onset of
anomalous degradation in IGBTs under power cycling
Summary:Presented the use of
statistical methods, including Principal Component Analysis (PCA) and a modified K-Nearest Neighbor technique,
to detect the anomalies of IGBTs in real-time.
- Prognostics and Health Management of High Power Light Emitting Diodes (LEDs)
Objective:Predict remaining useful life of LEDs using multiple
health indicators which can be directly obtained from
measurements
Summary:Presented a datadriven
method incorporating similarity-based statistical measures to estimate the remaining useful life of LEDs. The
proposed method is able to predict the remaining useful life of LEDs with unit-to-unit variations using multiple
health indicators that can be directly obtained from measurements.
- A Novel Warranty Servicing Approach based on Prognostics
Objective:• Develop a prognostics-based warranty framework to
increase the manufacturer’s marketability and profit and
reduce the customer’s life cycle cost
• Create a cost model for the prognostics-based warranty
to estimate the costs of the manufacturer and the
customer
Summary:Presented the use of PHM methods as a
proactive means to improve business warranties, in terms of identifying no-fault-found failures, understanding root
causes of failures, alleviating problems associated with recalls, enabling optimized warranty service and logistics
decisions, and creating new warranty strategies.
- Lifetime Warranty Policies for Electronic Products
Objective:• To review existing lifetime warranty policies and models
• To propose odifications to the existing models to adapt
them for modeling lifetime costs of electronic products
Summary:Modeled the lifetime of electronic products as a
function. The proposed method is able to devise an implementable lifetime warranty policy for electronic products
and develop the associated cost models.
- IEEE Standard for PHM
Summary:Beneficial to those who wish to implement prognostics for complex systems
through the development of an IEEE standard for PHM.
Past Research Projects
